Mistral Large 4 reportedly beats Qwen 3.8 Max and Kimi K3 on Terminal-Bench
Evermoving- · reddit · 2026-10-07
A Reddit post (with screenshot) claims Mistral Large 4 outperforms Qwen 3.8 Max and Kimi K3 on the Terminal-Bench coding-agent benchmark. Unverified — model names and scores have not been confirmed by official channels, but would mark a notable agentic-coding upset for Mistral if true.
More from Models
- Token-based pricing is strange: unpredictable, decoupled from value, misaligned incentives — amankhan · 2026-10-07
- JevBench splits leaderboard: open-weight models and API providers now ranked separately — airesearch12 · 2026-10-07
- Hot take: active params barely matter for cyber capability evals — RL env coverage is key — teortaxesTex · 2026-10-07
- Decagon launches Voice 3 with Chord voice model and duplex architecture for customer agents — Scobleizer · 2026-10-07
- JEV-9B, a Qwen3.5-based calibrated decision model, trends on Hugging Face — autotrust · 2026-10-07
- "Astra Pause Syndrome": steering may be making models go silent, OpenAI has a workaround — thursdai_pod · 2026-10-07